Determination of Transferable Lower-Bound Fracture Toughness from Small Specimens

Abstract

Abstract The master-curve (MC) enables fracture toughness to be measured in the ductile-to-brittle temperature (DBT) regime of ferritic steels using small specimens. However, MC application to structural components remains unclear in some areas. Vital issues of regulatory concern are the differences in the reference temperature T0 obtained from specimens of different sizes and shapes, the extension of the MC into the upper transition range, and the acceptability, or not, of a given level of failure probability to be used in the analysis of safety-relevant components [e.g., nuclear power reactor pressure vessel (RPV)]. Based on general theoretical considerations, the existence of a deterministic temperature-dependent lower bound of KIc or KJc is postulated. It can serve to provide conservative fracture toughness values as required in a screening analysis, which is usually the first step of a fracture mechanics analysis of a structural component. A simple method to determine this lower bound from the reference temperature T0 is presented. Furthermore, since testing of small specimens, like in precracked Charpy (PCC) tests, often leads to E1921-invalid data, an alternative procedure to estimate the MC is suggested that can use the invalid data. The practicability of the method is shown by a few examples. Based on the same method, a lower-bound of dynamic fracture toughness can be estimated from Charpy data sets. Combined with an empirical temperature shift, the lower bound of static fracture toughness can be estimated as well.